During the last 10 yrs it has been established that familial early onset AD is a genetically heterogeneous disease that can be caused by mutations in at least three different genes: Beta-amyloid protein precursor (APP)and presenilins 1 and 2 (1-3). Most mutations in these genes cause AD through changes in APP processing that elevate levels of total Beta-amyloid (A-Beta) or specifically increase A-Beta42 (4), providing strong support for the "Amyloid Hypothesis" of AD pathogenesis. Study of the genetics of late onset AD (LOAD) has also led to the identification of the epsilon 4 allele of the apolipoprotein E (APOE) gene as a risk factor. It has been shown to increase risk for AD in every population studied although the magnitude of the increase in risk varies between populations (6). Recent studies in animal models of A-Beta deposition have demonstrated that APOE4 also increases risk for AD through an A-Beta-related mechanism (7). To identify additional genetic risk factors for LOAD we performed a genome screen in 450 affected sibling pairs. The strongest evidence for linkage was obtained on chromosome 10 with a peak multipoint Iod score of 3.9. Younkin and colleagues have also provided evidence for a quantitative trait locus (QTL) that influences plasma A-Beta levels in the same region of chromosome 10 suggesting that the chromosome 10 AD locus may also influence AD risk via an A-Beta-dependent mechanism. Nine other chromosomal locations gave an MLS >1, including chromosome 12 where we and others have previously reported evidence of linkage. Although the linkage to chromosome 12 is modest, sib pair analyses with covariates increased the Iod score from 1.35 to 4.51 in APOE4 negative sib pairs. Furthermore a genome screen on the same dataset using age of onset as a quantitative trait also provided good evidence for a QTL influencing age of onset in the same region of chromosome 12. In the current application we propose to use linkage and linkage disequilibrium analyses to identify the disease alleles on chromosomes 10 and 12. We will use both case-control and family-based association methods to analyze individual single nucleotide polymorphisms (SNPs) and SNP haplotypes in genes under the linkage peak. SNPs will be identified from the public databases and by sequencing the genes in AD cases. Putative risk alleles/haplotypes will be tested for a functional effect on A-Beta levels in an in vitro model. Risk alleles will be characterized in several populations to determine their importance. [unreadable] [unreadable] [unreadable] [unreadable]